CN112346893A - Fault prediction method, device, terminal and storage medium - Google Patents
Fault prediction method, device, terminal and storage medium Download PDFInfo
- Publication number
- CN112346893A CN112346893A CN202011245611.4A CN202011245611A CN112346893A CN 112346893 A CN112346893 A CN 112346893A CN 202011245611 A CN202011245611 A CN 202011245611A CN 112346893 A CN112346893 A CN 112346893A
- Authority
- CN
- China
- Prior art keywords
- fault
- key component
- failure
- data
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 238000012423 maintenance Methods 0.000 claims description 27
- 238000013210 evaluation model Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 230000009467 reduction Effects 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000012847 principal component analysis method Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/008—Reliability or availability analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J9/00—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting
- H02J9/04—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source
- H02J9/06—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Emergency Management (AREA)
- Power Engineering (AREA)
- Business, Economics & Management (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention provides a fault prediction method, a fault prediction device, a terminal and a storage medium, wherein the method comprises the following steps: acquiring a plurality of fault factors of the key component and a plurality of historical fault time points of the key component; training by using a neural network according to a plurality of fault factors of the key component and a plurality of historical fault time points of the key component to obtain the predicted fault time of the key component; calculating the difference value between the current time and the predicted failure time of the key component, and predicting the failure possibility of the key component; the neural network is used for training, the expected failure time of the key component can be obtained, and the failure possibility of the key component is determined by analyzing the failure factor. After the method is applied, the characteristics of the key components can be refined into fault factors which are easy to control, the states of the key components are analyzed, the components of the target object to be damaged are determined, and the possibility of the target object to be in fault is predicted.
Description
Technical Field
The present invention relates to fault processing technologies, and in particular, to a fault prediction method, an apparatus, a terminal, and a storage medium.
Background
The power supply system of the UPS cooperating with the electrical system is widely applied to the data center to ensure the reliability of the data center. The core equipment of the mode is the UPS, so that the reliability of the UPS is improved, and the safety and the reliability of a power supply system cannot be fundamentally solved. Therefore, the UPS has been developed from a simple independent power supply device to a network device having a plurality of communication modes to manage seamless integration with the IT system.
However, since the development of new UPS's is too rapid, the overall structure is similar to a multi-part assembly, with some parts being susceptible to damage. Thus, the possibility of failure of the primary components of the UPS is anticipated to better ensure proper operation of the device.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: and predicting the device of the target object, and estimating the possibility of the main device of the target object to be in failure.
In order to solve the technical problems, the invention adopts the technical scheme that: provided is a failure prediction method, comprising the steps of: acquiring a plurality of fault factors of the key component and a plurality of historical fault time points of the key component; training by using a neural network according to a plurality of fault factors of the key component and a plurality of historical fault time points of the key component to obtain the predicted fault time of the key component; and calculating the difference value of the current time and the predicted failure time of the key component, and predicting the failure possibility of the key component.
Further, the difference value between the current time of each key component and the failure time is obtained and compared, and the key component with the damaged target object is determined.
Specifically, the number of hidden layers of the neural network is determined according to the working condition of the target object;
performing model training by using an MLP neural network, and establishing a state evaluation model of the key component;
and calculating the predicted failure time of the key component by applying the state evaluation model of the key component.
Further, segmenting the historical parameters of the fault factors to generate training set data, verification set data and test set data;
carrying out neural network training on the training set data, and establishing a state evaluation model of the key component;
applying the validation set data to adjust a state assessment model of the critical component;
and verifying the state evaluation model of the key component according to the test set data.
Acquiring operation data, maintenance data and fault records of the key components on a plurality of target objects in real time in a historical time period;
classifying and sorting the operation data, the maintenance data and the fault records according to the type of the target object to generate historical fault information of the key component characteristics;
selecting a plurality of fault factors of the key component and a plurality of historical fault time points of the key component from the historical fault information of the key component characteristics;
the operation data comprises use habit information, operation mode information and load regulation information of a user.
Specifically, a principal component analysis method is applied to perform dimensionality reduction processing on the historical fault information of the key component characteristics, and a plurality of fault factors of the key component are generated.
Further, unifying formats of the operation data, the maintenance data and the fault records according to the type of the target object;
acquiring threshold ranges of the operation data, the maintenance data and the fault records with uniform formats to eliminate data abnormal points;
and respectively smoothing the operation data, the maintenance data and the fault records with the abnormal points removed to obtain the data change trend of the target object.
The present application further provides a failure prediction apparatus, including:
the extraction module is used for acquiring a plurality of fault factors of the key component and a plurality of historical fault time points of the fault factors;
the training module is used for training by applying a neural network according to the plurality of fault factors of the key component and the plurality of historical fault time points of the fault factors to obtain the predicted fault time of the key component;
and the prediction module is used for calculating the difference value between the current time and the predicted failure time of the key component and predicting the failure possibility of the key component.
The present application further provides a terminal comprising a processor, a memory, and a display, the processor coupled to the memory and the display, the memory having stored thereon a computer program executable on the processor; the processor executes the computer program to realize the method.
The present application further provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described above
The invention has the beneficial effects that: the neural network is used for training, the expected failure time of the key component can be obtained, and the failure possibility of the key component is determined by analyzing the failure factor. After the method is applied, the characteristics of the key components can be refined into fault factors which are easy to control, the states of the key components are analyzed, the components of the target object to be damaged are determined, and the possibility of the target object to be in fault is predicted.
Drawings
The detailed structure of the invention is described in detail below with reference to the accompanying drawings
FIG. 1 is a flow chart of a method of fault prediction in an embodiment of the present invention;
FIG. 2 is a schematic diagram of neural network training using MLP in an embodiment of the present invention;
FIG. 3 is a block diagram of an apparatus for predicting a failure of a target object according to an embodiment of the present invention.
Detailed Description
In order to explain technical contents, structural features, and objects and effects of the present invention in detail, the following detailed description is given with reference to the accompanying drawings in conjunction with the embodiments.
Referring to fig. 1, fig. 1 is a flowchart illustrating a fault prediction method according to a first embodiment of the invention.
The invention provides a fault prediction method, which comprises the following steps:
s100, acquiring a plurality of fault factors of a key component and a plurality of historical fault time points of the key component;
s200, training by using a neural network according to a plurality of fault factors of the key component and a plurality of historical fault time points of the key component to obtain the predicted fault time of the key component;
step S300, calculating the difference value between the current time and the predicted failure time of the key component, and predicting the failure possibility of the key component;
the invention has the beneficial effects that: the neural network is used for training, the expected failure time of the key component can be obtained, and the failure possibility of the key component is determined by analyzing the failure factor. After the method is applied, the characteristics of the key components can be refined into fault factors which are easy to control, the states of the key components are analyzed, the components of the target object to be damaged are determined, and the possibility of the target object to be in fault is predicted.
The step S100 of acquiring a plurality of failure factors and a plurality of historical failure time points of the key component includes:
step S110, collecting operation data, maintenance data and fault records of a plurality of target objects in real time within preset time;
in this step, the target object device full life cycle big data monitoring platform can be used to collect the operation data, maintenance data and fault records of multiple target objects. The target object equipment full life cycle big data monitoring platform adopts a distributed centralized acquisition scheme, so that the target object equipment full life cycle big data monitoring platform can be transversely expanded or cut. In a further scheme, the platform can be used for performing ledger management on each device and establishing automatic device operation and maintenance record management. Therefore, the generation of an information isolated island can be avoided, data can be completely shared, and the target object can be conveniently subjected to deep excavation. The tracing of the fault equipment is realized, the product quality is objectively evaluated, and the data are applied to assist research and development, improve the product design and process in production, so as to improve the product quality.
Step S120, classifying and sorting the operation data, the maintenance data and the fault records according to the type of the target object to generate historical fault information of the target object;
step S130, historical fault information of key component characteristics, a plurality of fault factors and a plurality of historical fault time points of the key component are obtained.
It should be understood that the type of the target object may be classified according to the type of the target object, the operating condition of the target object, or a component of the target object.
The classification according to the types of the target objects is focused on maintaining the target objects sold by the customers and providing maintenance suggestions for the target objects used by the users. The method is classified according to the use environment of the target object, and is beneficial to researching the problem caused by the working condition of which target object; classification by target object with a component focuses on studying the component and on studying the different states of a component in different environments.
The operation data of the target object comprises use habit information, operation mode information and load adjustment information of a user. By applying the operation data of the target object, the whole operation process of the target object can be known, the analysis of the cause of the fault of the target object from the operation perspective is facilitated, and the fault possibly generated by the key component is predicted.
The maintenance data includes inspection data, maintenance data, and repair data. By applying the maintenance data, the behavior of the target object maintainer can be known, the level of the target object maintainer can be deduced, and the method is helpful for analyzing the reason of the target object fault and predicting the fault possibly generated by the key component from the perspective of the maintainer.
The fault records comprise fault information of the target object and fault data of each component of the target object. By using the fault records, the data of the target object can be conveniently recombined, analyzed and researched, and the fault reason can be better found from the perspective of a producer.
In addition, the operation data, the maintenance data and the fault records of the target object have certain redundancy, and the information is used to help eliminate uncertainty, so that the fault information can be acquired more accurately.
Specifically, step S120, in the step of sorting the operation data, the maintenance data, and the fault records according to the type of the target object, includes:
and step S121, unifying formats of the operation data, the maintenance data and the fault records according to the type of the target object.
It should be understood that the types of the target objects are various, and the target objects obtained by different classification methods may belong to different manufacturers or different series of the same manufacturer, so in this embodiment, the data may be formatted uniformly to facilitate processing between different types of data.
And S122, eliminating abnormal data points in the operation data, the maintenance data and the fault records.
It should be understood that, because data may have an inconsistency problem, some contradictions may exist between the operation data, the maintenance data and the fault records, and these data outliers may be directly eliminated in order to reduce the calculation difficulty.
And S123, smoothing the operation data, the maintenance data and the fault records respectively to obtain the data change trend of the target object.
It should be understood that the operation data, the maintenance data and the fault records are respectively smoothed, and the removed data abnormal points and the missing data are complemented by a mathematical method, so that the data can be used for neural network calculation, a plurality of data change trends of the target object are obtained, and the possibility of the fault of the key component of the target object is predicted. The present embodiment is a method based on data-driven prediction, and determines the data change trend of the target object through analysis of a large amount of data.
In a further embodiment, the step of obtaining the historical failure information of the target object in the above embodiment may be combined, or other manners may also be adopted to obtain the historical failure information of the target object.
In a specific embodiment, the step S130 of obtaining historical failure information of the key component, a plurality of failure factors and a plurality of historical failure time points of the key component includes:
and S131, performing dimensionality reduction on the historical fault information of the key component characteristics by applying a principal component analysis method to generate a plurality of fault factors of the key component.
Wherein the principal component analysis method is a dimension reduction analysis method. On the premise of keeping the main information of the original data, the problem of processing the multidimensional variables into a small number of comprehensive variables is solved, information overlapping among a plurality of variables can be effectively reduced, and the accuracy of an analysis result is improved.
It should be understood that, in the present embodiment, the dimension reduction of the historical failure information of the target object is performed based on the global consideration of the target object. In the embodiment, the dimension of the historical fault information of the target object is reduced, and some low-possibility events can be eliminated, so that the input elements required by the state evaluation model of the key component are greatly reduced, the calculation amount is saved, the diagnosis efficiency is enhanced, and meanwhile, the evaluation accuracy is still high.
Referring to fig. 2, fig. 2 is a schematic diagram of neural network training using MLP according to an embodiment of the present invention.
Specifically, step S200, training by using a neural network according to a plurality of fault factors of the key component and a plurality of historical fault time points of the key component, and obtaining a predicted fault time of the key component, including
And step S210, determining the number of hidden layers of the neural network according to the working condition of the target object.
In the step, the number of hidden layers of the neural network is set according to the complexity of the working condition of the target object, the first hidden layer represents the first extraction of the features, the second hidden layer represents the second extraction of the features extracted for the first time, and the like.
S220, performing model training by using an MLP neural network, and establishing and improving a state evaluation model of a key component;
and step S230, applying the state evaluation model of the key component to calculate the predicted failure time of the key component.
Therefore, the state evaluation model of the key component can be closer to the target object, and the state of the key component can be obtained more accurately.
Optionally, step S220 includes:
step S221, segmenting the historical parameters of the fault factors to generate training set data, verification set data and test set data
And S222, carrying out neural network training on the training set data, and establishing a state evaluation model of the key component.
And step S223, applying the verification set data and adjusting the state evaluation model of the key component.
Therefore, the weight of the factor can be obtained, and the state evaluation model of the key component can be established more conveniently.
In addition, some unconventional parameters can be added into the state evaluation model in the process of adjusting the fault training parameters, so that fault verification parameters which are more suitable for the target object are generated, and the state of the target object is more accurately suggested. The irregular parameters may include information that is unique to the target object, such as operating condition information of the target object, usage habit information of the user, inspection data, maintenance data, and repair data.
Through the technical scheme of steps S221 to S223 in this embodiment, the correlation between the failure factor of the key component and the device life can be found, and the state of the key component can be more accurately analyzed by applying the state evaluation model of the key component.
In the above, in step S300, the difference between the current time and the expected failure time of the critical component is calculated, and the failure probability of the critical component is predicted.
Specifically, a difference between the current time and the expected failure time of the critical component may be calculated, and the smaller this difference, the lower the failure probability of the critical component; when this difference is larger, the probability of failure of the critical component is higher.
Further, the method may be configured to perform failure prediction on a target object, where the target object is installed with at least two key components, and the method further includes:
and S400, acquiring and comparing the difference value between the current time and the expected failure time of each key component, and determining the key component with the damaged target object.
It will be appreciated that the smaller the difference between the current time and the expected failure time of a critical component, the lower its likelihood of failure.
To better illustrate the above rectifier failure prediction, the present invention uses one embodiment to illustrate it. To predict the state of the rectifier to an hour accuracy, the following method may be employed:
first, the rectifier fault factors are: voltage, temperature, PFC hardware protection signal, rectification IGBT stage C protection signal, input contactor contact anomaly signal, buffer contactor contact anomaly signal, which are marked as: x is the number of1,x2,x3,x4,x5,x6
Secondly, in the preset time, the failure time of the rectifier is 2018-02-0202: 02, 2018-08-0808: 08:08, 2018-10-1010: 10, 2019-01-0101: 01:01, 2019-06-0606: 06:06, 2019-09-0909: 09:09, 2020-03-0303: 03:03:
thus, in the last two years, the rectifier operating data at 5 minute intervals, shown in table 1, is:
| Time | x1 | x2 | x3 | x4 | x5 | x6 |
| 2018-01-01 00:00 | 236 | 11 | 0 | 0 | 0 | 0 |
| 2018-01-01 00:05 | 247 | 27 | 0 | 0 | 1 | 1 |
| 2018-01-01 00:10 | 217 | 30 | 0 | 1 | 0 | 0 |
| … | … | … | … | … | … | … |
| 2020-02-28 23:50 | 230 | 29 | 1 | 0 | 0 | 0 |
| 2020-02-28 23:55 | 228 | 18 | 0 | 0 | 0 | 0 |
TABLE 1
Applying the above table to establish a prediction model: the input required is x1,x2,x3,x4,x5,x6The output is the time of next failurey, establishing a model:
y＝f(x1,x2,x3,x4,x5,x6)
MLP neural networks were selected for model training, as shown in FIG. 2
Next, in a more specific embodiment, how to calculate the time of next failure based on historical data is explained:
at the time of 2018-01-0100: 00, the time point of next fault is 2018-02-0202: 02: 02;
then at time 2018-01-0100: 00, the next time it fails is 793.034 hours later.
From this calculation, a model-trained dataset is obtained, as in table 2:
TABLE 2
And inputting the data set into an MLP neural network model for training to obtain a rectifier fault prediction model.
If the current time is 2020-03-0510:10:10, the current x1,x2,x3,x4,x5,x6225, 22, 0, 0, 0, 0, input to the predictive model resulted in 730.03, due to 730.03>1, show that the rectifier can not be out of order in one hour in the future, if the input into the model obtains 0.88<1, which indicates that the rectifier is predicted to fail within one hour in the future.
Fig. 3 is a block diagram of a failure prediction apparatus for a target object according to an embodiment of the present invention. The present invention also provides a failure prediction apparatus for a target object, comprising:
the extraction module 100 is configured to obtain a plurality of fault factors of a key component and a plurality of historical fault time points of the fault factors;
the training module 200 is configured to apply a neural network to train according to the multiple fault factors of the key component and the multiple historical fault time points of the fault factors, and obtain a predicted fault time of the key component;
and the prediction module 300 is used for calculating the difference value between the current time and the predicted failure time of the key component and predicting the failure possibility of the key component.
It should be understood that the above modules are only roughly divided modules, and the various functions inside the modules can be formed by combining various units. In essence, the apparatus is a virtual execution subject for carrying the method.
The present application further provides a terminal comprising a processor, a memory, and a display, the processor coupled to the memory and the display, the memory having stored thereon a computer program executable on the processor; the processor executes the computer program and the computer program,
the invention also provides a storage medium comprising a computer program which, when loaded and executed by a processor, carries out the steps of the method as described above.
The processor may be a central analysis unit, but may also be other general purpose processors, digital signal processors, application specific integrated circuits, field programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method of fault prediction, the method comprising the steps of:
acquiring a plurality of fault factors of the key component and a plurality of historical fault time points of the key component;
training by using a neural network according to the plurality of fault factors of the key component and the plurality of historical fault time points of the key component to obtain the predicted fault time of the key component;
calculating the difference value between the current time and the predicted failure time of the key component, and predicting the failure possibility of the key component;
wherein the smaller the difference between the current time of the critical component and the expected failure time, the lower the failure probability of the critical component.
2. The failure prediction method of claim 1, wherein the method is used for failure prediction of a target object, the target object having at least two of the critical components installed, the method further comprising:
and acquiring and comparing the difference value between the current time of each key component and the failure time, and determining the damaged key component of the target object.
3. The method of predicting a fault of claim 1, wherein the step of training with a neural network to obtain the predicted time to failure of the critical component comprises:
determining the number of hidden layers of the neural network according to the working condition of the target object;
performing model training by using an MLP neural network, and establishing a state evaluation model of the key component;
and calculating the predicted failure time of the key component by applying the state evaluation model of the key component.
4. The method of fault prediction according to claim 3, wherein the step of applying the MLP neural network for model training to build a state estimation model of the critical component comprises:
segmenting the historical parameters of the fault factors to generate training set data, verification set data and test set data;
carrying out neural network training on the training set data, and establishing a state evaluation model of the key component;
applying the validation set data to adjust a state assessment model of the critical component;
and verifying the state evaluation model of the key component according to the test set data.
5. The fault prediction method of claim 1, wherein the step of obtaining a plurality of fault factors for the critical component and a plurality of historical fault time points for the critical component comprises:
collecting operation data, maintenance data and fault records of the key components on a plurality of target objects in real time in a historical time period;
classifying and sorting the operation data, the maintenance data and the fault records according to the type of the target object to generate historical fault information of the key component characteristics;
selecting a plurality of fault factors of the key component and a plurality of historical fault time points of the key component from the historical fault information of the key component characteristics;
the operation data comprises use habit information, operation mode information and load regulation information of a user.
6. The method of fault prediction according to claim 5, wherein said step of extracting a plurality of fault factors for said critical component from historical fault information for said critical component features comprises:
and performing dimensionality reduction on the historical fault information of the key component characteristics by applying a principal component analysis method to generate a plurality of fault factors of the key component.
7. The failure prediction method of claim 5, wherein the step of sorting the operation data, the maintenance data and the failure records according to the type of the target object comprises:
unifying formats of the operation data, the maintenance data and the fault record according to the type of the target object;
acquiring threshold ranges of the operation data, the maintenance data and the fault records with uniform formats to eliminate data abnormal points;
and respectively smoothing the operation data, the maintenance data and the fault records with the abnormal points removed to obtain the data change trend of the target object.
8. A failure prediction apparatus, comprising:
the extraction module is used for acquiring a plurality of fault factors of the key component and a plurality of historical fault time points of the fault factors;
the training module is used for training by applying a neural network according to the plurality of fault factors of the key component and the plurality of historical fault time points of the fault factors to obtain the predicted fault time of the key component;
and the prediction module is used for calculating the difference value between the current time and the predicted failure time of the key component and predicting the failure possibility of the key component.
9. A terminal comprising a processor, a memory, and a display, the processor coupled to the memory and the display, the memory having stored thereon a computer program executable on the processor; the processor executes the computer program to implement the method of one of claims 1 to 7.
10. A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of the method of one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011245611.4A CN112346893A (en) | 2020-11-10 | 2020-11-10 | Fault prediction method, device, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011245611.4A CN112346893A (en) | 2020-11-10 | 2020-11-10 | Fault prediction method, device, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112346893A true CN112346893A (en) | 2021-02-09 |
Family
ID=74363130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011245611.4A Pending CN112346893A (en) | 2020-11-10 | 2020-11-10 | Fault prediction method, device, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112346893A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949733A (en) * | 2021-03-12 | 2021-06-11 | 暨南大学 | Method and device for acquiring fault factor data, electronic device and storage medium |
| CN114611324A (en) * | 2022-03-28 | 2022-06-10 | 北京航空航天大学 | Fault diagnosis method for high-power magnetic plasma electric propulsion system |
| CN114881321A (en) * | 2022-04-29 | 2022-08-09 | 三一汽车起重机械有限公司 | Mechanical component failure prediction method, device, electronic device and storage medium |
| CN116260235A (en) * | 2023-03-11 | 2023-06-13 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109751206A (en) * | 2019-02-25 | 2019-05-14 | 宜通世纪物联网研究院(广州)有限公司 | fan blade icing failure prediction method, device and storage medium |
| CN110472288A (en) * | 2019-07-17 | 2019-11-19 | 武汉科技大学 | A kind of prediction technique, system, device and the medium of Bolt Torque decaying |
| US20200020186A1 (en) * | 2018-07-11 | 2020-01-16 | Acsys Holdings Limited | Systems and methods for providing an access management platform |
| CN111105063A (en) * | 2018-10-26 | 2020-05-05 | 北京国双科技有限公司 | Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium |
| CN111163484A (en) * | 2018-11-07 | 2020-05-15 | 中国移动通信集团湖南有限公司 | Base station fault prediction method and device |
| CN111506043A (en) * | 2020-04-09 | 2020-08-07 | 中国船舶工业综合技术经济研究院 | Fault prediction system for key components of naval gun |
| CN111581889A (en) * | 2020-05-26 | 2020-08-25 | 瑞纳智能设备股份有限公司 | Fault prediction method, system and equipment for heating equipment assembly |
-
2020
- 2020-11-10 CN CN202011245611.4A patent/CN112346893A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200020186A1 (en) * | 2018-07-11 | 2020-01-16 | Acsys Holdings Limited | Systems and methods for providing an access management platform |
| CN111105063A (en) * | 2018-10-26 | 2020-05-05 | 北京国双科技有限公司 | Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium |
| CN111163484A (en) * | 2018-11-07 | 2020-05-15 | 中国移动通信集团湖南有限公司 | Base station fault prediction method and device |
| CN109751206A (en) * | 2019-02-25 | 2019-05-14 | 宜通世纪物联网研究院(广州)有限公司 | fan blade icing failure prediction method, device and storage medium |
| CN110472288A (en) * | 2019-07-17 | 2019-11-19 | 武汉科技大学 | A kind of prediction technique, system, device and the medium of Bolt Torque decaying |
| CN111506043A (en) * | 2020-04-09 | 2020-08-07 | 中国船舶工业综合技术经济研究院 | Fault prediction system for key components of naval gun |
| CN111581889A (en) * | 2020-05-26 | 2020-08-25 | 瑞纳智能设备股份有限公司 | Fault prediction method, system and equipment for heating equipment assembly |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949733A (en) * | 2021-03-12 | 2021-06-11 | 暨南大学 | Method and device for acquiring fault factor data, electronic device and storage medium |
| CN112949733B (en) * | 2021-03-12 | 2022-06-07 | 暨南大学 | Method and device for acquiring fault factor data, electronic device and storage medium |
| CN114611324A (en) * | 2022-03-28 | 2022-06-10 | 北京航空航天大学 | Fault diagnosis method for high-power magnetic plasma electric propulsion system |
| CN114881321A (en) * | 2022-04-29 | 2022-08-09 | 三一汽车起重机械有限公司 | Mechanical component failure prediction method, device, electronic device and storage medium |
| CN116260235A (en) * | 2023-03-11 | 2023-06-13 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
| CN116260235B (en) * | 2023-03-11 | 2023-08-22 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112346893A (en) | Fault prediction method, device, terminal and storage medium | |
| KR101984730B1 (en) | Automatic predicting system for server failure and automatic predicting method for server failure | |
| CN112330034A (en) | Fault prediction method and system | |
| US20050288812A1 (en) | Quality prognostics system and method for manufacturing processes | |
| US10444746B2 (en) | Method for managing subsystems of a process plant using a distributed control system | |
| CN109992895B (en) | Method for extracting and predicting equipment performance degradation trend | |
| CN111796233B (en) | Method for evaluating secondary errors of multiple voltage transformers in double-bus connection mode | |
| CN113888353A (en) | Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment | |
| KR20090001148A (en) | Virtual metrology system and virtual metrology method | |
| CN116933197B (en) | Fault discrimination method and system for electricity consumption information acquisition system based on big data | |
| CN104335056A (en) | Interposer between a tester and material handling equipment to separate and control different requests of multiple entities in a test cell operation | |
| US9117045B2 (en) | System and method to predict chip IDDQ and control leakage components | |
| JP3718765B2 (en) | Plant diagnostic equipment | |
| CN113484817A (en) | Intelligent electric energy meter automatic verification system abnormity detection method based on TSVM model | |
| CN111832174A (en) | Wire loss rate processing method and device based on multiple regression | |
| CN112664410B (en) | Big data-based modeling method for unit online monitoring system | |
| CN115544803B (en) | Transformer residual life prediction method, device, equipment and storage medium | |
| CN115114124A (en) | Host risk assessment method and device | |
| KR101884907B1 (en) | Big Data Analytics Based Reliability Prediction Method | |
| CN116400266A (en) | Transformer fault detection method, device and medium based on digital twin model | |
| CN115640542A (en) | Intelligent electric energy meter state evaluation method and evaluation device based on Bayesian theory | |
| CN113887990A (en) | Electrical equipment maintenance decision optimization method | |
| CN113610225A (en) | Quality evaluation model training method and device, electronic equipment and storage medium | |
| CN114003466A (en) | Fault root cause positioning method for micro-service application program | |
| CN109492913B (en) | Modular risk prediction method and device for power distribution terminal and storable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |